Book Review – The Bell Curve
The Bell Curve
Intellignece and Class Structure in American Life
Richard Herrnstein and Charles Murray

It has been 11 years since this important book was published. It initially raised a firestorm of protest and controversy, though, as far as I know, none of its findings have been seriously challenged in the intervening years.

The Bell Curve is a very detailed scholarly review of what is known about the distribution of intelligence in America. As such it is primarily (as the title suggests) a book about statistics. It does not attempt to explain the biological roots of intelligence; it is not a neurology or psychology book. It rather deals with populations, and asks questions such as: Is intelligence an actual quantity that can be reliably measured? Does this quantity predict other useful things with any accuracy? How does this quantity vary between individuals and between groups?

The authors with painstaking references to the original data assert and support the following points:

• Intelligence (or cognitive ability) is not a human invention. It is a quantity that can be reliably measured, like blood pressure, or weight, and like those quantities it exists whether or not we measure it and is not a construct of a particular culture.


• Intelligence predicts many important things, and in many cases predicts them better than any other piece of information. IQ (Intelligence Quotient – an individual's score on an intelligence test), for example, predicts job performance in all sorts of jobs better than performance in an interview, better than educational achievement, and better than tests that try to measure job-specific skills. Meaning when hiring a cashier, a 12 minute IQ test would be more valuable than a test in operating a check-out scanner. Obviously, this (like many other points in the book) is a statistical argument, so there may be many individual exceptions while the general rule still holds.


• Americans with high IQ (called "the cognitive elite" by the authors) are more than ever before in our country's history becoming very well educated, earning high incomes, and working, marrying and living near other very intelligent people. It is very likely that such people never have substantive interactions with people of average or low IQ. The authors warn that we are becoming increasingly balkanized by IQ, and that technology, by allowing us to communicate and work without geographic barriers is allowing this to happen.


• Low IQ very strongly predicts living in poverty, having children out of wedlock, committing crime, being arrested, having a marriage end in divorce, having children that are neglected or abused and other behaviors associated with the social problems plaguing America. The often repeated scare that any of us, given an illness or a layoff, are a few paychecks away from poverty, is false.

  Among people who are both smart and well educated, the risk of poverty approaches zero.

• Any measurable characteristic that varies between individuals, whether it's IQ, or time in a 100 meter sprint, or hair color, or weight, is determined to some extent by heredity and to some extent by environment. IQ, for example, can be lowered by malnutrition or by extreme parental neglect – environmental effects. It is also true that high IQ parents tend to have high IQ children, and low IQ parents tend to have low IQ children. As our society has improved in raising the standard of living of all Americans, the environmental effects are playing a smaller role. Very few children (compared to 50 years ago) lose IQ because of an impoverished environment. Interestingly, this makes inheritance more important in determining IQ variability, as our environments become more equal.

  It is the central irony of egalitarianism: Uniformity in society makes the members of families more similar to each other and members of different families more different.

  Meaning, as society becomes more egalitarian, the genetic differences that run in families make a bigger difference in the variability between individuals. So more than ever, what determines how smart your kids are is how smart you are.


• Like any other quality that differs between individuals, IQ, on average, differs between ethnic groups. Though this fact may be upsetting, it is hard to imagine how it could not be so. We take it for granted that different ethnic groups, on average, have differing amounts of muscle mass, differing incidence of high blood pressure, diabetes and other diseases, and differing body proportions. At the same time we understand that within ethnic groups there is a wide variability of all of those characteristics between individuals. It would be astounding if intelligence were the one quality in which ethnic groups on average were identical. This is, of course, the part of the book that raised much controversy, although the authors go out of their way multiple times to stress that differences between the averages of ethnic groups tell us nothing about any specific individual, and that each person should be treated individually, not merely as a representative of a group. Again and again they stress the American ideal of individualism, and argue against prejudging individuals. Nevertheless, because they honestly presented what is known about race and intelligence, they were accused of racism. The authors analyze and dismiss the old accusation that IQ tests are racially or culturally biased, and give evidence why this can not be true. IQ predicts graduating from college, workplace income, and other important markers of success as well for African Americans as for whites. This would not be true if the test was biased against African Americans.

The authors then build on this foundational data by examining current affirmative action policies, both in education and in the workplace. They show that the magnitude of the advantage currently given to ethnic minorities is dramatically higher than what would be defensible given the initial stated goals of affirmative action, i.e. to assure that the institutions are casting a very wide net in encouraging applicants and that in cases in which two applicants have very similar qualifications, every effort is made to give the minority applicant the nod. Current affirmative action policies don't resemble these goals at all and have essentially devolved into two separate applications processes, one for whites (and non-protected minorities such as Asians) and another for protected minorities. These processes are entirely independent of each other and applicants from one pool are never compared to applicants in the other. The only goal is to assure that the mandated numbers of minority applicants are accepted, regardless of how low the qualifications of those applicants are, or how high those of the white applicants are. The statistical differences reported in the qualifications between accepted white and minority applicants are surprising – usually over a full standard deviation. The authors argue that, besides the obvious cost to the white (often poor white) applicants who lose the positions to much less qualified minority applicants, society in general pays other serious costs, including making minority achievement suspect, and lowering the economic efficiency of companies that are forced to hire much less productive employees.

In the last chapter the authors recommend remedies to the problems that they assert are caused by policies that ignore variability in intelligence. They do not recommend abandoning affirmative action, but instead suggest returning to the radical vision of individualism imagined by the Founding Fathers in which Americans are thought primarily as individuals. Then if small advantages must be given to groups that have been historically disadvantaged, they do not object, but this advantage must be small in comparison to the weight given the individual's achievements.

The book is very interesting, but some parts are a bit dry, and it is not short. Nevertheless I think it's must-reading for any who still support affirmative action, or who believe that the Federal Government's role in the last 50 years in education and employment has been positive. It's been 11 years. It would be nice to see some policymakers take it seriously. - served up by Doctor Bean @ 10:12 PM